STOCK PRICE PREDICTION

BACHELOR OF COMPUTER ENGINEERING

By

Sr. No. Name of the Student IEN No.

1. Aditya Yadav 12117030
2. Aditya Shelke 12127022
3. Sairaj Pawar 12127034
4. Jayant Singh 12127031
5. Ataharav Patil 12117025

Under the Guidance:

Bhagyashri Gawande

Department of Computer Science & Design Engineering

New Horizon Institute of Technology and Management Thane
University of Mumbai (2024-2025)

New Horizon Institute of Technology and Management,

Thane
 CERTIFICATE

This is to certify that the mini-project entitled (“Stock Price Prediciton”) of

Sr. No. Name of the Student IEN No.

1. Aditya Yadav 12117030

2. Aditya Shelke 12127022
3. Sairaj Pawar 12127034
4. Jayant Singh 12127031
5. Atharav Patil 12117025

has satisfactorily completed the mini-project in partial fulfillment of the

requirement for the award of the Degree of Bachelor of Engineering in
“Computer Science & Desgin ”.

Guide
Bhagyashri Gawande

2
 CONTENTS

Abstract.............................................................................................................................6

1.0 Introduction........................................................................................................... 7

2.0 Literature Survey.................................................................................................. 8

3.0 Problem Statement, Objectives and Scope .........................................................9
4.0 Proposed System ....................................................................................................11

5.0 Implementation Plan .............................................................................................13

6.0 Expected Outcome .................................................................................................14

6.1 : User interface 6.2 : Closing price vs Time chart 6.3 : Closing price vs

Time chart with 100 Days MA 6.4 : Closing price vs time chart with 100

Days MA and 200 Days MA 6.5 : Prediction Vs Original graph

7.0 References ...............................................................................................................17

8.0 Conclusion ..............................................................................................................18

5
 ABSTRACT

The project aims to develop a machine learning-based approach for predicting

stock prices with the objective of improving investment decisions. Stock
prediction is a complex task due to the many factors that influence stock prices,
such as economic indicators, company news, global events, and investor
sentiment. Therefore, the project proposes an approach that integrates
mathematical functions, machine learning algorithms, and external factors to
achieve higher prediction accuracy and profitability in trading.

The project recognizes two types of stocks: intraday trading and long-term
investment. Intraday trading, also known as day trading, is a popular form of trading
where traders buy and sell securities within a single trading day, usually aiming to
make a profit on short-term price movements. In contrast, long-term investors buy
stocks with the intention of holding them for a more extended period, typically
months or years, with the expectation of earning returns from price appreciation and
dividends.

The project employs Long Short-Term Memory (LSTM) models, a type of neural
network, which are highly effective in sequence prediction tasks. LSTM models
can retain information about previous data points and use it to inform predictions,
making them useful for predicting stock prices. In particular, past stock prices are
essential for predicting future stock prices, and LSTM models can leverage this
information to make predictions.

CHAPTER 1

INTRODUCTION

In today's world, the stock market plays a crucial role in the global economy. The
stock market is a complex system that involves the buying and selling of stocks or

6
shares of a company, which represents a small portion of the ownership in that
company. With the advancements in technology, it has become easier than ever to
access and analyze stock market data.
This project aims to provide an in-depth analysis of the stock market using various
data visualization and analysis techniques. The project will involve collecting
historical stock market data for various companies and analyzing them to identify
trends, patterns, and correlations. The data will be visualized using charts, graphs,
and other visual aids to help investors and traders make informed decisions.
The project will also explore different techniques for predicting future stock prices
using machine learning algorithms such as regression analysis, LSTM. The results of
these predictions will be visualized and compared to actual stock prices to evaluate
the effectiveness of the prediction models.
Overall, this project aims to provide a comprehensive understanding of the

stock market and its trends, as well as to provide insights that can help investors
and traders make informed decisions.
(1.1) Purpose

Stock analysis is also called equity analysis or market analysis. Investors or

traders make buying or selling decisions based on stock analysis information.
Stock analysis helps traders to gain an insight into the economy, stock market, or
securities.

CHAPTER 2
LITERATURE SURVEY

"A comprehensive study of stock market prediction techniques in Indian financial

market" by N. K. Soni and P. Sharma (2021) - This survey provides an overview of
recent research on stock market prediction techniques in the Indian financial market.
It covers various methods, including machine learning, statistical models, and
technical analysis, and their effectiveness in predicting stock prices. The survey also

7
discusses the challenges and limitations of using such techniques in the Indian
context.

"A review of Indian stock market prediction using machine learning

techniques" by R. Bhardwaj and R. Sharma (2021) - This literature survey focuses on

the use of machine learning techniques for stock market prediction in the Indian
context. It covers various methods, such as neural networks, decision trees, and
support vector machines, and their effectiveness in predicting stock prices. The
survey also addresses the challenges and limitations of using machine learning for
stock analysis in India.

CHAPTER 3 PROBLEM STATEMENT,

OBJECTIVES AND SCOPE

(3.1) Problem Statement

One of the main problems regarding stock analysis is the difficulty in predicting future stock
prices accurately. Stock prices are influenced by a wide range of factors, including
economic conditions, company performance, industry trends, and global events, which can
make it challenging to identify and analyze all the relevant information accurately.
Another problem is the availability and quality of data. Stock analysis requires access
to reliable and accurate financial data, such as company financial statements, market
trends, and economic indicators. However, such data can be difficult to obtain, and
its quality can vary significantly, making it challenging to analyze and draw
meaningful conclusions.
Finally, stock analysis is a complex and dynamic process that requires a high level of
expertise and experience. Many factors can influence stock prices, and successful analysis
requires a thorough understanding of financial markets, investment strategies, and
analytical tools, which can be challenging to master.

8
(3.2) Objectives
1. To Understand Historical Performance: Visualization and analysis tools
allowinvestors to view a stock's historical price movements and trading volume
over time. This information helps investors to understand how the stock has
performed in the past, identify any patterns or trends, and make informed
decisions about buying or selling the stock.

2. To Identify Market Trends: By analyzing stock market data, investors

canidentify market trends and potential opportunities. Visualization tools can
help investors to visualize trends in the stock market, such as upward or
downward movements in stock prices, changes in trading volume, and other
indicators.
3. To Analyze Company Performance: Visualization and analysis tools
canalso help investors to understand the financial performance of a company,
including revenue, profits, and other key financial indicators. By analyzing these
factors, investors can make informed decisions about the potential future growth
of the company and the potential return on investment.
4. To Manage Risk: Visualization and analysis tools can also helpinvestors to
manage risk by identifying potential risks and opportunities. By analyzing
historical and current market data, investors can identify potential risks and
make informed decisions about whether to invest in a particular stock.
5. To Make Informed Decisions: Ultimately, the objective of stock
visualization and analysis is to help investors make informed decisions about
buying and selling stocks. By providing a comprehensive view of the market and
the individual stocks, investors can make informed decisions that are based on
data and analysis rather than speculation and intuition.

(3.3) Scope
The scope of stock prediction refers to the ability to use data and analysis to forecast the
future performance of a particular stock or the stock market as a whole. This can include

9
predicting changes in stock prices, volatility, and trends based on various factors such as
economic indicators, company financial statements, and market sentiment.
There are different approaches to stock prediction, ranging from fundamental
analysis, which looks at the financial health of a company, to technical analysis,
which focuses on patterns in stock price charts.
The scope of stock prediction is significant for investors, traders, and financial
analysts who use this information to make investment decisions. Accurate stock
prediction can help investors maximize returns, minimize risks, and identify
opportunities in the stock market. However, it's important to note that stock
prediction is not foolproof and involves a degree of uncertainty and risk.

CHAPTER 4 PROPOSED SYSTEM

10
Data Collection: The first step will involve collecting historical stock market
data from various sources such as financial news websites, stock exchange
websites, and other data providers. This data will be collected in a structured
format and stored in a database.
Data Preprocessing: The collected data will be preprocessed to remove any
missing values, outliers, and errors. The data will be transformed into a suitable
format for analysis and visualization. Data
Visualization: The preprocessed data will be visualized using various
visualization techniques such as line charts, bar charts, candlestick
charts, and heat maps. These visualizations will help to identify trends,
patterns, and correlations in the data.

11
Data Analysis: The visualized data will be analyzed using statistical and
machine learning techniques to identify patterns, correlations, and
anomalies in the data. This analysis will provide insights into the stock
market and help to make informed decisions.
Prediction Models: Machine learning algorithms such as regression analysis,
neural networks, and decision trees will be used to develop prediction models
for future stock prices. These models will be trained using historical data and
evaluated using testing data. The accuracy of these models will be measured
and compared. User Interface: The system will provide a user-friendly
interface for users to interact with the data and visualizations. Users will be
able to explore the data, customize visualizations, and make predictions using
the prediction models.
Deployment: The system will be deployed on a web server or cloud platform
to make it accessible to users. The system will be scalable to handle a large
amount
CHAPTER 5
IMPLEMENTATION PLAN

12
 Chart Title
#REF! 9- Feb 15- Feb 8- Mar 13- Mar 16- Apr
21- Feb 2-Mar

Documentation

Adjustments

Test and
Reviews
Development
Phase
Sourcing

Preparatory
phase

0 10 20 30 40 50 60 70

CHAPTER 6 EXPECTED
OUTCOME

(6.1) user Interface :

13
(6.2) Closing Price VS Time Chart :

14
(6.3) Closing Price VS Time Chart with 100Days MA :

(6.4) Closing Price VS Time Chart with 100Days MA & 200Days MA :

15
(6.5) Predictions VS Original :

CHAPTER 7
REFERENCE

[1] Brownlee, J. (2019). “Time Series Forecasting with Python ” MachineLearning

Mastery.

[2] Chen, J., Li, X., & Li, Y. (2019). “Machine learning for stock investment:

A s[3y]s tGemonaztiacl erze,v Jie. wA”. ,I &EE WEo Aocdcse, sDs, .7 D, 5. 5(29091-85)5.
m45a, rket data: A review.” Journal of Visual Languages & Computing,

48-67.
[4] Hochreiter, R., & Schmidbauer, H. (2018) , “Stock price prediction

using

16
onee p learning”. In Proceedings of the 2018 International Joint Conferenced Neural
Networks (IJCNN) (pp. 1-8). IEEE.

CHAPTER 8 CONCLUSION

(8.1) Conclusion
The stock visualization and analysis project proposes a comprehensive platform for
analyzing and visualizing stock market data. It involves several stages, including
data collection, preprocessing, visualization, analysis, prediction models, user
interface, and deployment. The project aims to provide valuable insights into the
stock market to help investors and traders make informed decisions. It explores
various machine learning algorithms, such as regression analysis, neural networks,
and decision trees, to develop prediction models for future stock prices. Overall, the
project is an essential tool for anyone interested in the stock market, providing a
user-friendly interface, visualization techniques, and prediction models to maximize
returns.

(8.2) Future Scope

The future scope of stock prediction is vast, as advancements in technology and data
analytics are making it easier to analyze and interpret large amounts of financial and
market data.
One area of potential growth in stock prediction is the use of artificial intelligence
(AI) and machine learning algorithms to analyze financial data and identify
patterns and trends. This can help investors and analysts

Informed decisions by predicting future market movements based on past

performance and other factors.

17